Multiple Feature Extraction of Image using 2D CA

Feature Extraction is a long active topic in image processing, and it attracts lots of attention over the last five decades due to its uses in real-life applications. Edge detection method allows users to discover the features of an image for a notable change in the gray level. This texture indicates an ending of one region and the beginning of another in the image. Likewise, over the last few decades, researchers have been exploring to exploit the simple computing model of Cellular Automata (CA) with local neighborhood structure in image processing techniques. The usage of CA in applications of medical image-processing is identified by Kendall Preston [1] in 1979. It is found that edge detection using CA has a potential advantage over traditional approaches for its lightweight nature and computational efficiency. This paper introduces a new technique for image edge detection using two-dimensional CA. The method can use specific rule sets of CA using the Von Neumann neighborhood for edge detection with null boundary conditions. The performance analysis of the scheme is done and compared with some existing standard edge detection techniques. The results obtained from the proposed technique is very promising for feature extraction.